Resource Allocation for Cognitive Network Slicing in PD-SCMA System Based on Two-Way Deep Reinforcement Learning

In this paper,we propose the Two-way Deep Reinforcement Learning(DRL)-Based resource allocation algorithm,which solves the problem of resource allocation in the cognitive downlink network based on the underlay mode.Secondary users(SUs)in the cognitive network are multiplexed by a new Power Domain Sparse Code Multiple Access(PD-SCMA)scheme,and the physical resources of the cognitive base station are virtualized into two types of slices:enhanced mobile broadband(eMBB)slice and ultrareliable low latency communication(URLLC)slice.We design the Double Deep Q Network(DDQN)network output the optimal codebook assignment scheme and simultaneously use the Deep Deterministic Policy Gradient(DDPG)network output the optimal power allocation scheme.The objective is to jointly optimize the spectral efficiency of the system and the Quality of Service(QoS)of SUs.Simulation results show that the proposed algorithm outperforms the CNDDQN algorithm and modified JEERA algorithm in terms of spectral efficiency and QoS satisfaction.Additionally,compared with the Power Domain Non-orthogonal Multiple Access(PD-NOMA)slices and the Sparse Code Multiple Access(SCMA)slices,the PD-SCMA slices can dramatically enhance spectral efficiency and increase the number of accessible users.

An Intelligent Admission Control Scheme for Dynamic Slice Handover Policy in 5G Network Slicing

5G use cases,for example enhanced mobile broadband(eMBB),massive machine-type communications(mMTC),and an ultra-reliable low latency communication(URLLC),need a network architecture capable of sustaining stringent latency and bandwidth requirements;thus,it should be extremely flexible and dynamic.Slicing enables service providers to develop various network slice architectures.As users travel from one coverage region to another area,the callmust be routed to a slice thatmeets the same or different expectations.This research aims to develop and evaluate an algorithm to make handover decisions appearing in 5G sliced networks.Rules of thumb which indicates the accuracy regarding the training data classification schemes within machine learning should be considered for validation and selection of the appropriate machine learning strategies.Therefore,this study discusses the network model’s design and implementation of self-optimization Fuzzy Qlearning of the decision-making algorithm for slice handover.The algorithm’s performance is assessed by means of connection-level metrics considering the Quality of Service(QoS),specifically the probability of the new call to be blocked and the probability of a handoff call being dropped.Hence,within the network model,the call admission control(AC)method is modeled by leveraging supervised learning algorithm as prior knowledge of additional capacity.Moreover,to mitigate high complexity,the integration of fuzzy logic as well as Fuzzy Q-Learning is used to discretize state and the corresponding action spaces.The results generated from our proposal surpass the traditional methods without the use of supervised learning and fuzzy-Q learning.

Attention-based neural network for end-to-end music separation

The end-to-end separation algorithm with superior performance in the field of speech separation has not been effectively used in music separation.Moreover,since music signals are often dual channel data with a high sampling rate,how to model longsequence data and make rational use of the relevant information between channels is also an urgent problem to be solved.In order to solve the above problems,the performance of the end-to-end music separation algorithm is enhanced by improving the network structure.Our main contributions include the following:(1)A more reasonable densely connected U-Net is designed to capture the long-term characteristics of music,such as main melody,tone and so on.(2)On this basis,the multi-head attention and dualpath transformer are introduced in the separation module.Channel attention units are applied recursively on the feature map of each layer of the network,enabling the network to perform long-sequence separation.Experimental results show that after the introduction of the channel attention,the performance of the proposed algorithm has a stable improvement compared with the baseline system.On the MUSDB18 dataset,the average score of the separated audio exceeds that of the current best-performing music separation algorithm based on the time-frequency domain(T-F domain).

End-to-End Auto-Encoder System for Deep Residual Shrinkage Network for AWGN Channels

With the rapid development of deep learning methods, the data-driven approach has shown powerful advantages over the model-driven one. In this paper, we propose an end-to-end autoencoder communication system based on Deep Residual Shrinkage Networks (DRSNs), where neural networks (DNNs) are used to implement the coding, decoding, modulation and demodulation functions of the communication system. Our proposed autoencoder communication system can better reduce the signal noise by adding an “attention mechanism” and “soft thresholding” modules and has better performance at various signal-to-noise ratios (SNR). Also, we have shown through comparative experiments that the system can operate at moderate block lengths and support different throughputs. It has been shown to work efficiently in the AWGN channel. Simulation results show that our model has a higher Bit-Error-Rate (BER) gain and greatly improved decoding performance compared to conventional modulation and classical autoencoder systems at various signal-to-noise ratios.

Resting-state brain network remodeling after different nerve reconstruction surgeries:a functional magnetic resonance imaging study in brachial plexus injury rats

Distinct brain remodeling has been found after different nerve reconstruction strategies,including motor representation of the affected limb.However,differences among reconstruction strategies at the brain network level have not been elucidated.This study aimed to explore intranetwork changes related to altered peripheral neural pathways after different nerve reconstruction surgeries,including nerve repair,endto-end nerve transfer,and end-to-side nerve transfer.Sprague–Dawley rats underwent complete left brachial plexus transection and were divided into four equal groups of eight:no nerve repair,grafted nerve repair,phrenic nerve end-to-end transfer,and end-to-side transfer with a graft sutured to the anterior upper trunk.Resting-state brain functional magnetic resonance imaging was obtained 7 months after surgery.The independent component analysis algorithm was utilized to identify group-level network components of interest and extract resting-state functional connectivity values of each voxel within the component.Alterations in intra-network resting-state functional connectivity were compared among the groups.Target muscle reinnervation was assessed by behavioral observation(elbow flexion)and electromyography.The results showed that alterations in the sensorimotor and interoception networks were mostly related to changes in the peripheral neural pathway.Nerve repair was related to enhanced connectivity within the sensorimotor network,while end-to-side nerve transfer might be more beneficial for restoring control over the affected limb by the original motor representation.The thalamic-cortical pathway was enhanced within the interoception network after nerve repair and end-to-end nerve transfer.Brain areas related to cognition and emotion were enhanced after end-to-side nerve transfer.Our study revealed important brain networks related to different nerve reconstructions.These networks may be potential targets for enhancing motor recovery.

Research on Data Privacy Protection Algorithm with Homomorphism Mechanism Based on Redundant Slice Technology in Wireless Sensor Networks

Wireless transmission method in wireless sensor networks has put forward higher requirements for private protection technology. According to the packet loss problem of private protection algorithm based on slice technology, this paper proposes the data private protection algorithm with redundancy mechanism, which ensures privacy by privacy homomorphism mechanism and guarantees redundancy by carrying hidden data. Moreover,it selects the routing tree generated by CTP(Collection Tree Protocol) as routing path for data transmission. By dividing at the source node, it adds the hidden information and also the privacy homomorphism. At the same time,the information feedback tree is established between the destination node and the source node. In addition, the destination node immediately sends the packet loss information and the encryption key via the information feedback tree to the source node. As a result,it improves the reliability and privacy of data transmission and ensures the data redundancy.

Incentive Scheme for Slice Cooperation Based on D2D Communication in 5G Networks

In the 5th generation(5G)wireless communication networks,network slicing emerges where network operators(NPs)form isolated logical slices by the same cellular network infrastructure and spectrum resource.In coverage regions of access points(APs)shared by slices,device to device(D2D)communication can occur among different slices,i.e.,one device acts as D2D relay for another device serving by a different slice,which is defined as slice cooperation in this paper.Since selfish slices will not help other slices by cooperation voluntarily and unconditionally,this paper designs a novel resource allocation scheme to stimulate slice cooperation.The main idea is to encourage slice to perform cooperation for other slices by rewarding it with higher throughput.The proposed incentive scheme for slice cooperation is formulated by an optimal problem,where cooperative activities are introduced to the objective function.Since optimal solutions of the formulated problem are long term statistics,though can be obtained,a practical online slice scheduling algorithm is designed,which can obtain optimal solutions of the formulated maximal problem.Lastly,the throughput isolation indexes are defined to evaluate isolation performance of slice.According to simulation results,the proposed incentive scheme for slice cooperation can stimulate slice cooperation effectively,and the isolation of slice is also simulated and discussed.

SDN Based Next Generation Mobile Network With Service Slicing and Trials

Software-Defined Network （SDN） empowers the evolution of Internet with the OpenFlow, Network Virtualization and Service Slicing strategies. With the fast increasing requirements of Mobile Internet services, the Internet and Mobile Networks go to the convergence. Mobile Networks can also get benefits from the SDN evolution to fulfill the 5th Generation （5G） capacity booming. The article implements SDN into Frameless Network Architecture （FNA） for 5G Mobile Network evolution with proposed Mobile-oriented OpenFlow Protocol （MOFP）. The Control Plane/User Plane （CP/UP） separation and adaptation strategy is proposed to support the User-Centric scenario in FNA. The traditional Base Station is separated with Central Processing Entity （CPE） and Antenna Element （AE） to perform the OpenFlow and Network Virtualization. The AEs are released as new resources for serving users. The mobile-oriented Service Slicing with different Quality of Service （QoS） classification is proposed and Resource Pooling based Virtualized Radio Resource Management （VRRM） is optimized for the Service Slicing strategy with resource-limited feature in Mobile Networks. The capacity gains are provided to show the merits of SDN based FNA. And the MiniNet based Trial Network with Service Slicing is implemented with experimental results.

Enabling IoT Network Slicing with Network Function Virtualization

Numerous Internet of Things (IoT) devices are being connected to the net-works to offer services. To cope with a large diversity and number of IoT ser-vices, operators must meet those needs with a more flexible and efficient net-work architecture. Network slicing in 5G promises a feasible solution for this issue with network virtualization and programmability enabled by NFV (Net-work Functions Virtualization). In this research, we use virtualized IoT plat-forms as the Virtual Network Functions (VNFs) and customize network slices enabled by NFV with different QoS to support various kinds of IoT services for their best performance. We construct three different slicing systems including: 1) a single slice system, 2) a multiple customized slices system and 3) a single but scalable network slice system to support IoT services. Our objective is to compare and evaluate these three systems in terms of their throughput, aver-age response time and CPU utilization in order to identify the best system de-sign. Validated with our experiments, the performance of the multiple slicing system is better than those of the single slice systems whether it is equipped with scalability or not.

Generating Questions Based on Semi-Automated and End-to-End Neural Network

With the emergence of large-scale knowledge base,how to use triple information to generate natural questions is a key technology in question answering systems.The traditional way of generating questions require a lot of manual intervention and produce lots of noise.To solve these problems,we propose a joint model based on semi-automated model and End-to-End neural network to automatically generate questions.The semi-automated model can generate question templates and real questions combining the knowledge base and center graph.The End-to-End neural network directly sends the knowledge base and real questions to BiLSTM network.Meanwhile,the attention mechanism is utilized in the decoding layer,which makes the triples and generated questions more relevant.Finally,the experimental results on SimpleQuestions demonstrate the effectiveness of the proposed approach.

Tibetan Multi-Dialect Speech Recognition Using Latent Regression Bayesian Network and End-To-End Mode

We proposed a method using latent regression Bayesian network (LRBN) toextract the shared speech feature for the input of end-to-end speech recognition model.The structure of LRBN is compact and its parameter learning is fast. Compared withConvolutional Neural Network, it has a simpler and understood structure and lessparameters to learn. Experimental results show that the advantage of hybridLRBN/Bidirectional Long Short-Term Memory-Connectionist Temporal Classificationarchitecture for Tibetan multi-dialect speech recognition, and demonstrate the LRBN ishelpful to differentiate among multiple language speech sets.

Efficient Network Slicing with Dynamic Resource Allocation

With the rapid development of wireless network technologies and the growing de⁃mand for a high quality of service(QoS),the effective management of network resources has attracted a lot of attention.For example,in a practical scenario,when a network shock oc⁃curs,a batch of affected flows needs to be rerouted to respond to the network shock to bring the entire network deployment back to the optimal state,and in the process of rerouting a batch of flows,the entire response time needs to be as short as possible.Specifically,we re⁃duce the time consumed for routing by slicing,but the routing success rate after slicing is re⁃duced compared with the unsliced case.In this context,we propose a two-stage dynamic net⁃work resource allocation framework that first makes decisions on the slices to which flows are assigned,and coordinates resources among slices to ensure a comparable routing suc⁃cess rate as in the unsliced case,while taking advantage of the time efficiency gains from slicing.

Downlink Resource Allocation for NOMA-Based Hybrid Spectrum Access in Cognitive Network

To solve the contradiction between limited spectrum resources and increasing communication demand,this paper proposes a wireless resource allocation scheme based on the Deep Q Network(DQN)to allocate radio resources in a downlink multi-user cognitive radio(CR)network with slicing.Secondary users(SUs)are multiplexed using non-orthogonal multiple access(NOMA).The SUs use the hybrid spectrum access mode to improve the spectral efficiency(SE).Considering the demand for multiple services,the enhanced mobile broadband(eMBB)slice and ultrareliable low-latency communication(URLLC)slice were established.The proposed scheme can maximize the SE while ensuring Quality of Service(QoS)for the users.This study established a mapping relationship between resource allocation and the DQN algorithm in the CR-NOMA network.According to the signal-to-interference-plusnoise ratio(SINR)of the primary users(PUs),the proposed scheme can output the optimal channel selection and power allocation.The simulation results reveal that the proposed scheme can converge faster and obtain higher rewards compared with the Q-Learning scheme.Additionally,the proposed scheme has better SE than both the overlay and underlay only modes.

Satellite E2E Network Slicing Based on 5G Technology

We investigate the design of satellite network slicing for the first time to provide customized services for the diversified applications,and propose a novel scheme for satellite end-to-end(E2E) network slicing based on 5G technology,which provides a view of common satellite network slicing and supports flexible network deployment between the satellite and the ground.Specifically,considering the limited satellite network resource and the characteristics of the satellite channel,we propose a novel satellite E2E network slicing architecture.Therein,the deployment of the network functions between the satellite and the ground is coordinately considered.Subsequently,the classification and the isolation technologies of satellite network sub-slices are proposed adaptively based on 5G technology to support resource allocation on demand.Then,we develop the management technologies for the satellite E2E network slicing including slicing key performance indicator(KPI) design,slicing deployment,and slicing management.Finally,the analysis of the challenges and future work shows the potential research in the future.

User Scheduling and Slicing Resource Allocation in Industrial Internet of Things

Heterogeneous base station deployment enables to provide high capacity and wide area coverage.Network slicing makes it possible to allocate wireless resource for heterogeneous services on demand.These two promising technologies contribute to the unprecedented service in 5G.We establish a multiservice heterogeneous network model,which aims to raise the transmission rate under the delay constraints for active control terminals,and optimize the energy efficiency for passive network terminals.A policygradient-based deep reinforcement learning algorithm is proposed to make decisions on user association and power control in the continuous action space.Simulation results indicate the good convergence of the algorithm,and higher reward is obtained compared with other baselines.

Demand Prediction Based Slice Reconfiguration Using Dueling Deep Q-Network

To satisfy diversified service demands of vertical industries,network slicing enables efficient resource allocation of a common infrastructure by creating isolated logical networks.However,uncertainty and dynamics of service demands will cause performance degradation.Due to operation costs and resource constraints,it is challenging to maintain high quality of user experience while obtaining high revenue for service providers(SPs).This paper develops an optimal and fast slice reconfiguration(OFSR)framework based on reinforcement learning,where a novel scheme is proposed to offer optimal decisions for reconfiguring diverse slices.A demand prediction model is proposed to capture changes in resource requirements,based on which the OFSR scheme is triggered to determine whether to perform slice reconfiguration.Considering the large state and action spaces generated from uncertain service time and resource requirements,deep dueling architecture is adopted to improve the convergence rate.Extensive simulations validate the effectiveness of the proposed framework in achieving higher long-term revenue for SPs.

A statistical end-to-end performance model for networks with complex topologies

Network calculus provides new tools for performance analysis of networks, but analyzing networks with complex topologies is a challenging research issue using statistical network calculus. A service model is proposed to characterize a service process of network with complex topologies. To obtain closed-form expression of statistical end-to-end performance bounds for a wide range of traffic source models, the traffic model and service model are expanded according to error function. Based on the proposed models, the explicit end-to-end delay bound of Fractional Brownian Motion（FBM） traffic is derived, the factors that affect the delay bound are analyzed, and a comparison between theoretical and simulation results is performed. The results illustrate that the proposed models not only fit the network behaviors well, but also facilitate the network performance analysis.

Machine Learning for Network Slicing Resource Management:A Comprehensive Survey

The emerging technology of multi-tenancy network slicing is considered as an es sential feature of 5G cellular networks.It provides network slices as a new type of public cloud services and therewith increases the service flexibility and enhances the network re source efficiency.Meanwhile,it raises new challenges of network resource management.A number of various methods have been proposed over the recent past years,in which machine learning and artificial intelligence techniques are widely deployed.In this article,we provide a survey to existing approaches of network slicing resource management,with a highlight on the roles played by machine learning in them.

Tackling IoT Scalability with 5G NFV-Enabled Network Slicing

With emerging large volume and diverse heterogeneity of Internet of Things (IoT) applications, the one-size-fits-all design of the current 4G networks is no longer adequate to serve various types of IoT applications. Consequently, the concepts of network slicing enabled by Network Function Virtualization (NFV) have been proposed in the upcoming 5G networks. 5G network slicing allows IoT applications of different QoS requirements to be served by different virtual networks. Moreover, these network slices are equipped with scalability that allows them to grow or shrink their instances of Virtual Network Functions (VNFs) when needed. However, all current research only focuses on scalability on a single network slice, which is the scalability at the VNF level only. Such a design will eventually reach the capacity limit of a single slice under stressful incoming traffic, and cause the breakdown of an IoT system. Therefore, we propose a new IoT scalability architecture in this research to provide scalability at the NS level and design a testbed to implement the proposed architecture in order to verify its effectiveness. For evaluation, three systems are compared for their throughput, response time, and CPU utilization under three different types of IoT traffic, including the single slice scaling system, the multiple slices scaling system and the hybrid scaling system where both single slicing and multiple slicing can be simultaneously applied. Due to the balanced tradeoff between slice scalability and resource availability, the hybrid scaling system turns out to perform the best in terms of throughput and response time with medium CPU utilization.

End-to-End Joint Multi-Object Detection and Tracking for Intelligent Transportation Systems

Environment perception is one of the most critical technology of intelligent transportation systems(ITS).Motion interaction between multiple vehicles in ITS makes it important to perform multi-object tracking(MOT).However,most existing MOT algorithms follow the tracking-by-detection framework,which separates detection and tracking into two independent segments and limit the global efciency.Recently,a few algorithms have combined feature extraction into one network;however,the tracking portion continues to rely on data association,and requires com‑plex post-processing for life cycle management.Those methods do not combine detection and tracking efciently.This paper presents a novel network to realize joint multi-object detection and tracking in an end-to-end manner for ITS,named as global correlation network(GCNet).Unlike most object detection methods,GCNet introduces a global correlation layer for regression of absolute size and coordinates of bounding boxes,instead of ofsetting predictions.The pipeline of detection and tracking in GCNet is conceptually simple,and does not require compli‑cated tracking strategies such as non-maximum suppression and data association.GCNet was evaluated on a multivehicle tracking dataset,UA-DETRAC,demonstrating promising performance compared to state-of-the-art detectors and trackers.